Saturday, July 18, 2015

Coffee - Different Roasts, Different Health Effects | Darker Roasts Better for Weight Loss, Lighter Roasts Healthier!?

Dark, medium or light roast - this is more than just a question of taste! The significant effect of roasting on the phenol-composition of coffee and the consequent health effects are rarely discussed.
You may remember my often misinterpreted article about bullet-proof coffee in which I actually defended the fad by providing evidence that the negative effects on blood lipids may actually be cause by not filtering the coffee vs. putting butter into it (read it). Well, today I would like to dig a bit deeper into the differential effects not of coffee brewing, but rather of coffee roasting on the of your favorite beverage's major active ingredients and its subsequent effects on your health.

Now, to get the bad news out of the way, the study from the West-German Centre of Diabetes and Health in Düsseldorf which evaluated these effects unfortunately did so in 118 overweight subjects.
You can learn more about coffee at the SuppVersity

Remember: With Coffee More Won't Help More

Coffee - The Good, Bad & Interesting

Three Cups of Coffee Keep Insulin At Bay

Caffeine's Effect on Testosterone, Estrogen & SHBG

The Coffee³ Ad- vantage: Fat loss, Appetite & Mood

Caffeine Resis- tance - Does It Even Exist?
The good news, however, is that the subjects who were recruited by articles in local media in the region of Düsseldorf, Germany, were mostly overweight, not obese (BMIs of ≥27 kg/m²) came from various age-groups (18–69 years old) and - this is important - were regular coffee consumers (at least 3 cups/day), just like you (?). With them being weight stable and free of any acute of chronic diseases (with the exception of type 2 diabetes mellitus), they are thus not the ideal study group, but decent subjects - even though we will have to be careful when it comes to making conclusions with regard to the health effects of differently roasted coffee for leaner and healthier people.

In the introduction I already alluded to the purpose of this study. What I still have to do, though, is to tell you exactly what the scientists wanted to know and how they tried to find the answer to their question(s). From previous studies, Kempf et al. knew that dark roast coffee, which is rich in the trigenolline byproduct N-methylpyridinium (NMP: 785 mmol/L) and low in chlorogenic acid metabolites has significantly more pronounced weight loss effects in pre-obese (but not normal-weight) subjects; see Figure 1) than light roast coffee when it is administered at doses of 500ml/day in a 2x4-week cross over study (Kotyczka. 2011).
Figure 1: Only the pre-obese subjects lost significant amounts of weight. A trend for higher greater weight loss in the dark roast phase of the cross-over study is however evident in all subjects (Kotcyzka. 2011).
Now, 4 weeks is not an acute study, but it's not exactly long enough to observe things like ceiling effects or the well-known "too much of a good thing" effect that may occur when agents with proven short-term benefits are administered over long(er) time periods.
A previously discussed problem with coffee is the formation of mold toxins, in particular ochratoxin A, when it's stored inappropiately. Luckily, roasting does - irrespective of whether you roast dark or light - sing. reduce the toxic mold (Van der Stegen. 2011).
What does the observational / epidemiological evidence say? Well, you know the problems with this type of data. Not only is it unreliable, it's also often not differentiated enough. It's thus impossible to tell whether the proven beneficial long-term effects of coffee consumption on type II diabetes risk (Greenberg. 2006; Muley. 2012; Natella. 2012 | you have to drink 3-5 cups per day) would occur with both the medium and dark roasts, but if we rely solely on the study at hand and the previously reported filtered- vs. unfiltered coffee (this is also scientifically confirmed based on observational studies | Muley. 2012) data, using medium roast coffee and a filter appears to be the most promising candidate for long-term health benefits from coffee consumption. This hypothesis is also warranted in view of the fact that the initial increase in total antioxidants that occurs with light roasting is reversed as the beans darken (del Castillo. 2002).
Against that background, you'd probably agree with Kerstin Kempf and her colleagues who thought it would be interesting to see "whether these effects on body weight would also be detectable after a longer-term consumption of dark roast coffee, in particular, in overweight adults" who are at the highest risk of dietary induced metabolic deteriorations (Kempf. 2015).
Figure 2: Flow sheet of recruitment and analysis (directly from Kempf. 2015)
In their randomized controlled trial, Kempf et al. compared the effect of 3-months consumption of (1) a medium roasted coffee blend rich in CQA and trigonelline versus a dark-roasted coffee blend rich in NMP (see Table 1 for the concentration of CQA, which is the sum of 3-, 4- and 5-caffeoylquinic acids, chlorogenic acid metabolites; NMP N-methylpyridinium, trigonelline and caffeine), as well as of (2) both groups combined for an estimate of general effects of coffee on body weight and selected cardiometabolic risk factors like HDL, LDL, triglycerides, adipokines (leptin, adiponectin), etc..
Table 1: Composition of medium roast (M)-coffee and dark roast (D)-coffee blends (Kempf. 2015).
What about filtered vs. unfiltered? The coffee in the Kempf study was handed to the subjects in regular commercially available coffee pads which look like coffee bagged in a filter. It is thus not surprising that similarly pronounced alterations in blood lipids as I've reported them before for unfiltered coffee, did not occur with either of the two roasts that were used in the study at hand.
During the study, the participants consumed an average of 4–5 cups per day of the roast they had been randomly assigned to. Even though that's probably more than the 500ml in the previously cited study by Kotcyzka et al. (see Figure 1) this was not enough coffee to exert significant effects on either the mean body weight, the body mass index or the waist circumference of the participants; and that's irrespective of whether the they consumed the dark (D) or medium (M) roast.
Figure 3: In contrast to the short-term cross-over study, the long(er) term study suggests that lighter roasts have the more beneficial effects on potentially CVD- (HDL + trigs) and T2DM (Trigs + Adiponectin) serum markers (Kempf. 2015).
Due to the different effects on some health-relevant metabolic markers, though, the study results are still SuppVersity news-worthy: In spite of the fact that the systolic blood pressure, which is obviously also a health relevant maker, decreased in the dark roast coffee group only (p < 0.05), other changes, in particular the contrast between
  • the significant increases in HDL and adiponectin, both of which have been linked to lower CVD and type II diabetes (T2DM) risk, which occurred in the medium roast group, and
  • the likewise significant increase in triglycerides, of which scientists believe that they pave the way for both diabetes and heart disease, which occurred only in the dark roast group,
would suggest health benefits of choosing lighter over darker roasts. How relevant those are for someone with perfect cholesterol and triglyceride levels may yet be questionable, also because neither of the two roasts affected the subjects' glucoregulation, or insulin levels (the scientists did observe an increase in the long-term glucose measure HbA1c (+0.1%), but this increase was identical in both groups and so marginal that it's hardly physiologically relevant).
If you look at the correlations between serum NMP levels (remember NMP is higher in dark roasts) the study confirms what the headline says: While darker roasts may favor weight loss (even though not necessarily sign. weight loss), it appears to have negative effects on leptin's healthy anti-diabetic and anti-CVD cousin adiponectin (Kempf. 2015).
So what? In contrast to the initially referenced study by Kotcyzka et al., the study at hand appears to confirm what previous studies on chlorogenic acid (CGA) already suggested (Thom. 2007; Onakpoya. 2015): Higher contents of CGA and/or its metabolites appear to be something to look for in both coffee extracts and regular coffee powder, because of their potential beneficial effects on blood glucose and CVD-markers.

The medium roast does after all not just contain more CGA metabolites, it does also have lower amounts of N-methylpyridinium, of which the study at hand clearly indicates that it is not just associated with weight loss (cf. Kotyczka, as well), but also w/ reduced adiponectin levels and overall less favorable effects on the few health-relevant changes the scientists observed during the study. It remains speculative, however, if said effects would have been different in (a) lean individuals (unlikely) or (b) subjects who don't belong to the ever-increasing group of regular coffee consumers (more likely), because the initial wash-out period during which the overweight subjects had to abstain from the consumption of coffee, cocoa, black or green tea probably didn't turn the coffee aficianados into caffeine naive coffee abstainers | Comment on Facebook!
  • Crozier, Thomas WM, et al. "Espresso coffees, caffeine and chlorogenic acid intake: potential health implications." Food & function 3.1 (2012): 30-33.
  • del Castillo, María Dolores, Jennifer M. Ames, and Michael H. Gordon. "Effect of roasting on the antioxidant activity of coffee brews." Journal of Agricultural and Food Chemistry 50.13 (2002): 3698-3703.
  • Greenberg, James A., Carol N. Boozer, and Allan Geliebter. "Coffee, diabetes, and weight control." The American journal of clinical nutrition 84.4 (2006): 682-693.
  • Kempf, Kerstin, et al. "Cardiometabolic effects of two coffee blends differing in content for major constituents in overweight adults: a randomized controlled trial." European journal of nutrition (2014): 1-10.
  • Muley, Arti, Prasad Muley, and Monali Shah. "Coffee to reduce risk of type 2 diabetes?: a systematic review." Current diabetes reviews 8.3 (2012): 162-168.
  • Natella, Fausta, and Cristina Scaccini. "Role of coffee in modulation of diabetes risk." Nutrition reviews 70.4 (2012): 207-217.
  • Nicoli, M. C., et al. "Antioxidant properties of coffee brews in relation to the roasting degree." LWT-Food Science and Technology 30.3 (1997): 292-297.
  • Kotyczka, Christine, et al. "Dark roast coffee is more effective than light roast coffee in reducing body weight, and in restoring red blood cell vitamin E and glutathione concentrations in healthy volunteers." Molecular nutrition & food research 55.10 (2011): 1582-1586.
  • Onakpoya, I. J., et al. "The effect of chlorogenic acid on blood pressure: a systematic review and meta-analysis of randomized clinical trials." Journal of human hypertension 29.2 (2015): 77-81.
  • Thom, Erling. "The effect of chlorogenic acid enriched coffee on glucose absorption in healthy volunteers and its effect on body mass when used long-term in overweight and obese people." Journal of International Medical Research 35.6 (2007): 900-908.
  • Van der Stegen, Gerrit HD, Paulus JM Essens, and Joost Van der Lijn. "Effect of roasting conditions on reduction of ochratoxin A in coffee." Journal of Agricultural and Food Chemistry 49.10 (2001): 4713-4715.